sign in sign up
<<
Home , Hypervitaminosis, Hypervitaminosis A

Gut, 1992, 33, 707-710 707 A absorption in cystic fibrosis: risk ofhypervitaminosis A Gut: first published as 10.1136/gut.33.5.707 on 1 May 1992. Downloaded from

D R James, G Owen, I A Campbell, M C Goodchild

Abstract state, using a high dose (approximately 7000 IU/ status was examined in nine adult kg body weight) vitamin A absorption test cystic fibrosis patients and six adult control without pancreatic enzyme supplements." subjects, together with an assessment of their In this study the absorption of 10000 IU ability to absorb 10 000 IU of retinyl palmitate vitamin A was examined in a group of adult from a test meal, taken with appropriate pan- cystic fibrosis patients who took a test meal, and creatic enzyme supplements. Median baseline pancreatic enzyme supplements as they judged values for plasma and carotene, as well appropriate. as median serum retinol binding protein con- centrations, were significantly lower in cystic fibrosis patients than in control subjects. One Methods cystic fibrosis patient had a raised fasting plasma retinyl ester concentration suggestive SUBJECTS AND SAMPLES of chronic A, but no Nine adult cystic fibrosis patients (seven men; symptoms of . Measures of vitamin A median age 22 years (range 16-29)) and six absorption were also significantly lower in control subjects (three men; median age 27 years cystic fibrosis patients, although there was (range 19-30)) took part in the study. All cystic considerable overlap with control values. No fibrosis patients had been taking regular vitamin correlation was observed between measures of A supplements and all but one used pancreatic baseline status and vitamin A absorption. enzyme supplements regularly. Measurement of plasma retinyl esters may be After an overnight fast (cystic fibrosis patients an appropriate investigation in those patients had been asked to omit vitamin supplements considered to be at risk of chronic hyper- from mid-day on the day before the test) samples

vitaminosis A. were taken for the assessment of vitamin A http://gut.bmj.com/ status, plasma retinol and retinyl esters, retinol binding protein, and prealbumin. Samples were Pancreatic exocrine insufficiency is usual in also obtained for total plasma carotenes, C cystic fibrosis' 2 and it is not surprising that low reactive protein, and their biochemical profile serum concentrations of fat soluble , (albumin, aspartate transaminase, y glutamyl particularly A and E, have been observed.3 Low transferase, alkaline phosphatase, and

serum vitamin A concentrations may not be the bilirubin). on September 28, 2021 by guest. Protected copyright. result of malabsorption alone, as increased A test meal consisting of 30 g cornflakes, 200 concentrations of vitamin A with low serum ml whole cow's milk, 120 g white bread, 20 g concentrations were reported by Underwood strawberry jam, 5 g sugar, and 10 000 IU vitamin and Denning in 1972.4 Even with oral vitamin A (as retinyl palmitate, supplied by Hoffman La- supplements, however, serum concentrations in Roche, UK) in 1 ml arachis oil was eaten with cystic fibrosis patients remain below those of pancreatic enzyme supplementation where control populations.4s appropriate. Hourly samples for plasma retinyl Chronic may exist for ester estimation were taken over the following six many years among members of the general hours. All subjects were fasted during the study, public without recognition,6 resulting eventually but allowed free access to water. Cystic fibrosis in hepatocellular damage and hepatic fibrosis.7 patients also completed a questionnaire detailing Pre-existing may also affect the regular medication, bowel function, and symp- capacity of the organ to transport the vitamin toms of hypo/hypervitaminosis A, as described Departments of Medical into the thus its elsewhere.`" 12 were also asked to undertake Biochemistry and Child circulation, potentiating hepa- They Health (Cystic Fibrosis totoxicity. As cystic fibrosis is a condition pre- a Pancreolauryl test, again using whatever pan- Unit), University disposing to liver pathology, it is not certain that creatic supplements they considered necessary. Hospital of Wales, supplementation of all patients with vitamin A is Cardiff D R James advisable. Doses of vitamin A need not be large G Owen to cause toxicity, as this has occurred with oral MATERIALS M C Goodchild intakes of 20-45 000 IU/day.5 Raised plasma Retinol, retinyl acetate, and retinyl palmitate Department of Chest retinyl ester concentrations and biochemical were obtained from Sigma Chemical Company, Diseases, Llandough evidence of liver damage have been reported in Poole, UK, and purified as necessary by alumina Hospital, Cardiff normal control subjects taking moderate doses column chromatography. All other chemicals I A Campbell only (10 000 IU or less daily) of the vitamin.9 and solvents were obtained from BDH Correspondence to: and were Analar or Dr D R James, Department of It has been recommended that cystic fibrosis Chemicals Ltd, Poole, UK, Medical Biochemistry, patients be supplemented with 10-25000 IU/ high performance liquid chromatography University Hospital of Wales, Cardiff. day,"' but little is known of how well these (HPLC) grade as appropriate. Plasma retinol Accepted for publication supplements are absorbed. Previous studies have was analysed by an isocratic reverse phase HPLC September 1991 examined the absorption ofvitamin A in a fasting method, based on those previously described by 708 J3ames, Owen, Campbell, Goodchild

TABLE Biochemical data in control subjects and cysticfibrosis patients takingpart in vitamin A absorption study Control subjects Cysticfibrosis patients 1 2 3 4 5 6 Median 1 2 3 4 5 6 7 8 9 Median Gut: first published as 10.1136/gut.33.5.707 on 1 May 1992. Downloaded from Age (yrs) 30 30 24 27 19 27 27 21 22 18 16 29 23 22 25 24 22 Serum albumin (g/l) 46 44 42 47 45 43 45 41 40 43 42 39 40 39 42 43 41* Serum aspartate transaminase (IU/l) 32 24 28 28 16 29 25 103 12 42 56 47 22 19 18 21 22 Serum glutamyl transferase (IU/l) 18 11 9 15 6 11 10 113 26 61 21 92 13 25 16 20 23* Serum alkaline phosphatase (IU/l) 79 48 48 80 59 86 69 214 90 164 354 195 99 109 166 124 164* Serum bilirubin (,umol/l) 10 9 10 12 17 17 11 19 5 9 10 21 10 5 12 5 10 Plasmaretinol([tmol/l) 2-39 1-72 1-14 1-62 1-59 1-59 1-61 1-23 0 93 1 07 1-23 0-86 1-37 1-00 0 65 1.19 1-04* Plasma retinol esters (nmol/l) 65 80 21 21 30 48 39 28 17 17 26 0 37 11 0 303 17 Serum retinol (mg/l) 50 40 26 40 41 34 38 28 25 28 37 23 33 25 22 35 28t Serum PA (mg/1) 317 314 165 316 269 234 292 204 186 205 238 223 208 168 169 262 205 Plasmacarotenes([tmol/1) 2-67 3 53 2-60 2 54 1-82 2.53 2.57 - 0 34 0.40 2-30 0-36 0-29 0.35 0-29 0-91 0.36* Daily vitamin A supplements (IU/dayx103) ------16 2-5 5 5 13 13 8 8 41 8 ARetinol esters (nmol/l) 160 345 361 78 520 477 353 - 70 145 206 64 22 172 47 133 266 133t Area under curve (nmol.l. '.h)1088 1368 1115 237 1681 650 1102 193 203 625 193 67 437 145 330 625 203* Lipasewithtestmeal(IUxlO3) ------30 20 10 8 104 32 0 15 112 20 Pancreolauryltest result (%) ------6 11 7 6 / 7 / 3 7 *p<0.01 compared with controls; tp<005 compared with controls: Mann-Whitney U test.

McClean et all4 and Bieri et al.` Plasma retinyl None of the patients reported symptoms con- esters were determined by a modification of the sistent with hypo- or hypervitaminosis A. The method of Bankson et al. 16 Serum retinol binding median daily quantity of vitamin A supplements protein was determined by Mancini radial taken regularly was 8000 IU (range 2500-41 000 immunodiffusion using commercially available IU); all but one patient (patient 9, who took plates and standards (Behring Diagnostics, 41 000 IU/day) took 16000 1U/day or less Hounslow, UK), and prealbumin by rate (Table). The median lipase dose (in the form of nephelometry (Beckman Auto-ICS) with Creon, Pancrease, Nutrizyme, or Cotazyme) Beckman reagents and standards. Plasma taken with the test meal was 20000 IU (range carotenes were measured spectrophotometric- 0-112000 IU). Some abnormalities of liver ally.'7 Samples for the Pancreolauryl test were function were seen among the cystic fibrosis analysed in accordance with the manufacturer's patients, who also showed lower concentrations

instructions. of plasma retinol, carotenes, and serum retinol http://gut.bmj.com/ binding protein (Table). No significant differ- ence was observed between the two groups for STATISTICS fasting plasma retinyl ester concentrations Results were analysed by Spearman's rank order (Table). correlation test or Mann-Whitney U test with Measurement of plasma retinyl ester concen- correction for ties, as appropriate. trations after the test meal showed that absorp-

tion profiles were very variable in both control on September 28, 2021 by guest. Protected copyright. subjects (Fig 1) and cystic fibrosis patients (Fig Results 2). The results of the absorption test are also There was no significant difference in the age presented as the maximum rise in plasma retinyl distribution of the patients and control subjects esters (A RE) and the area under the curve (Table). Serum C reactive protein concentra- (AUC), the baseline being taken as the fasting tions were normal (<6 mg/l) in six of nine cystic value (Fig 3A and B). The median values in fibrosis patients and not significantly raised 10-9 mg/l) in the remaining three. 600 0 600. E 0 E C~~~~~~~~~~~~ , 400 0~~~~~~~~~~~~~

C ~~~~~~~~~~~0 ) U) C.)~~~ ~ ~ ~ ~ ~ ~ ~~~~3

~200 -~ ~ ~ ~ ~ ~ ~ ~ ~~ 0

E E 0 ,v, 0 2 4 6 CL0 . . . . .Time (h) 0 2 4 6 X = Patient 9 Time (h) Figure 2: Vitamin A absorption profiles in cvsticfibrosis Figure 1: Vitamin A absorption profiles in control subjects. patients. Vitamin A absorption in cysticfibrosis: risk ofhypervitaminosis A 709

cystic fibrosis patients were significantly lower results (<20%, Table). These results, in con- than in control subjects, (p'0O05 for A RE and junction with low plasma carotene values p'O-01 for AUC), although there was consider- obtained in eight of nine patients (Table),

able overlap. No correlations were found indicate that it is very likely that all patients had Gut: first published as 10.1136/gut.33.5.707 on 1 May 1992. Downloaded from between measures of vitamin A absorption and significant malabsorption. All patients had a fasting plasma retinol or retinyl esters. clinical requirement for pancreatic enzyme One patient (patient 9), who had been taking supplements, although these were not well 41 000 IU vitamin A/day for at least one year and tolerated (and not taken) by one patient used large quantities of pancreatic enzyme (patient 7). supplements with meals (112 000 IU lipase with the test meal), had a fasting plasma retinyl ester concentration of 303 nmol/l, almost four times Discussion greater than that seen in any of the controls Retinol is absorbed by a carrier mediated passive (Table). Measures of his vitamin A absorption absorption process under physiological condi- (A RE and AUC) were the highest in the cystic tions, and by simple passive absorption at fibrosis group. However, circulating retinol and pharmacological concentrations." Absorption is retinol binding protein concentrations were in dependent on a number of factors, including the keeping with those of the other cystic fibrosis presence of pancreatic enzymes and bile salts. patients. He was a very fit patient who showed no Esterification of the newly absorbed retinol by clinical symptoms ofhypervitaminosis A, and no acyl-Co A: retinyl acyl transferase within the evidence clinically or biochemically of liver dys- enterocyte is reported to be normal in cystic function. fibrosis. 9 The newly esterified vitamin A is then Seven patients (including patient 9) undertook transported to the liver in association with the Pancreolauryl test, and all had abnormal chylomicron remnants.20 Observations on baseline vitamin A status of cystic fibrosis patients in this study are similar to 600 A those reported in other studies.4` Few data are available on vitamin A absorption in cystic 0 fibrosis, but in our study of nine patients, all 0 E of whom had pancreatic insufficiency, some 400 absorbed significant quantities of the preformed vitamin while others did not. It seems that in CO 0 some patients, mechanisms for vitamin A absorption are not compromised. This wide http://gut.bmj.com/ _ 0 a) variation in absorption ofvitamin A is in keeping 200, with observations reported recently on faecal E 0 losses of the vitamin in cystic fibrosis." CL results from the ingestion of -1 0 large quantities ofthe vitamin over a short period of time, leading to the appearance of free retinol

(that is, not bound to retinol binding protein) in on September 28, 2021 by guest. Protected copyright. CF Controls the circulation. Persistent consumption of 2000 smaller quantities of vitamin A over a prolonged B period may result in chronic hypervitaminosis A, with an increase in circulating vitamin A and in the ratio of ester to forms.22 The con- 0 siderably raised fasting plasma retinyl ester con- 1600 centration of 303 nmol/l seen in patient 9 (who was taking 41000 IU/day) suggests chronic hypervitaminosis A, although the patient was chronic 0 entirely asymptomatic. Traditionally, :2E 1200 hypervitaminosis A has been associated with an intake of larger quantities of the vitamin

0 + (>100000 IU/day) over a prolonged period. However, the daily intake required to produce -0 toxicity is very variable, ranging from 5000 IU'2

C_ 800 to 1 400 000 lU,24 and the length of time required WU is also variable." 6 It has also been suggested that a 0 there is an inherited variability of susceptibility to vitamin A toxicity.27 The toxicity of vitamin A may be modulated 400- 0 by a number of factors, being potentiated by 0 alcohol20 and protein-energy malnutrition29; a 0 degree of protection is afforded by tocopherol, , and zinc."' In cystic fibrosis the abnor- mality of vitamin A metabolism seems to be 0 related to an inability of the liver to mobilise its CF Controls stores of the vitamin, and perhaps it is not Figure 3: Measures ofvitamin A absorption in cvstic fibrosis patients and control subjects. (A) Maximum rise in plasma surprising that the patient with raised plasma retinol esters: p

normal plasma retinol concentration. This may 5 Knoepfler G, Rotthauwe HW, Odenthal A. Vitamin A, carotin, retinolbindendes protein und praealbumin in serum be analogous to a report of a non-cystic fibrosis von patienten mit cystischer fibrose. Z Kinderheilkunde patient who had liver damage from massive 1975; 119: 279-91. 6 Stimtson WH. Vitamin A intoxication in adults. Report of a

vitamin A deposition, together with a depressed case with a summary of the literature. N Engli] Med 1961; Gut: first published as 10.1136/gut.33.5.707 on 1 May 1992. Downloaded from serum vitamin A concentration. He had con- 265: 369-73. 7 Russell RM, Boyer JL, Bagheri SA, Hruban Z. Hepatic injury sumed up to 25 000 IU/day of the vitamin for from chronic hypervitaminosis A resulting in portal hyper- seven years and had a degree of protein-energy tension and ascites. N Engli] Med 1974; 291: 435-40. 8 Leo MA, Lieber CS. Hypervitaminosis A: a liver lover's as a result of a poor .29 lament. Hepatology 1988; 8: 412-7. Despite having the highest vitamin A absorp- 9 Krasinski SD, Russell RM, Otradovec CL, Sadowski JA, Hartz SE, Jacob RA, et al. Relationship of vitamin A and tion parameters, patient 9 ranked only 4th, 3rd, intake to fasting plasma retinol, retinol-binding and 2nd for plasma retinol, carotenes, and serum protein, retinyl esters, carotene, a-tocopherol; and choles- terol among elderly people and young adults: increased retinol binding proteins concentrations respect- plasma retinyl esters among vitamin A supplement users. ively. Animal studies suggest that over supple- Am]7 Clin Nutr 1989; 49: 112-20. 10 Guide to drug therapy in patients with cystic fibrosis. The mentation may also suppress retinol binding Cystic Fibrosis Foundation. 1972: 50-2. proteins synthesis,3' thus further complicating 11 Warwick WJ, Hausen LG, Sharp H. Absorption of vitamin A in patients with cystic fibrosis. Clin Pediatr 1976; 15: 807- the situation. 10. Our results suggest that care should be taken 12 Underwood BA. Vitamin A in animal and . In: Spom MB, Roberts AB, Goodman DS, eds. The . with preformed vitamin A supplementation as Vol 1. London: Academic Press, 1984: 338-42. some cystic fibrosis patients may be at risk of 13 Koerner WF, Voelim J. New aspects of the tolerance of retinol in humans. Intl Vitam Nutr 1975; 45: 363-72. developing hypervitaminosis A if oversupple- 14 McClean SW, Warwick WJ, Ruddell ME, Gross EG, De mented. Giovanna JJ, Pack GL. Liquid chromatographic assay for retinol (vitamin A) and retinol analogues in therapeutic The lower plasma carotene concentrations trials. Clin Chim Acta 1982; 28: 693-14. seen in the cystic fibrosis patients suggest that 15 Bieri JG, Tolliver TJ, Catignani GL. Simultaneous determ- ination of a-tocopherol and retinol in plasma or red cells by preformed vitamin A supplementation is neces- high pressure liquid chromatography. Am]7 Clin Nutr 1979; sary, as these patients may not be able to meet 32: 2143-9. 16 Bankson DD, Russell RM, Sadowski JA. Determination of their daily requirements for this vitamin from retinyl esters and retinol in serum or plasma by normal- dietary carotene alone. phase liquid chromatography: method and applications. Clin Chem 1986; 32: 35-40. The susceptibility of the liver to injury by 17 Gowenlock AH, ed. In: Varley's practical clinical chemistry. 6th increased stores of vitamin A in cystic fibrosis ed. London, Heineman. 1988: 895-9. 18 Hollander D. Intestinal absorption of vitamin A, E, D and K. patients, expecially those with liver dysfunction, JLab Clin Med 1981; 97: 449-62. is not known. It seems prudent therefore to 19 Rasmussen M, Michaelson H, Lie SO, Nilsson A, Petersen LB, Norum KR. Intestinal retinol esterification of serum assess in some way the extent of a patient's retinol in children with cystic fibrosis. ] Pediatr Gastroenterol vitamin A stores, but unfortunately no simple Nutr 1986; 5: 397-403. 20 Goodman DS, Huang HS, Shiratori T. Tissue distribution and non-invasive technique is available yet. Periodic metabolism of newly absorbed vitamin A in the rat. ] Lipid http://gut.bmj.com/ estimation of fasting plasma retinyl esters seems Res 1965; 6: 390-6. 21 Ahmed F, Ellis J, Murphy J, Wootton S, Jackson AA. to be the only option available at present; Excessive faecal losses of vitamin A (retinol) in cystic however, the plasma retinyl ester concentration fibrosis. Arch Dis Child 1990; 65: 589-93. 22 Smith FR, Goodman DS. Vitamin A transport in human may be raised only when liver stores ofvitamin A vitamin A toxicity. N EnglI] Med 1976; 294: 805-8. are saturated. Further difficulties in interpreta- 23 Schurr D, Herbert J, Habibi E, Abrahamov A. Case report: unusual presentation of vitamin A intoxication. ] Paediatr tion may occur as plasma ester concentrations Gastroenterol Nutr 1983; 2: 705-7.

may be raised as a consequence of liver path- 24 Fleischmann R, Schlote W, Schomerus H, Wolburg H, on September 28, 2021 by guest. Protected copyright. Castrillon-Oberndoffer WL, Hoensch H. Kleinotige ology. 6 We are currently investigating the poss- Leberzirorrhose mit ausgeprater portaler hypertension als ible use of other vitamin A metabolites as Folge einer Vitamin-A Intoxication bei Psoriasis- Behandlung. Dtsch Med Wochenschr 1977; 102: 1637-40. markers ofhepatic vitamin A stores. 25 Bush ME, Dahms BB. Fatal hypervitaminosis in neonate. Arch Pathol Lab Med 1984; 108: 838-42. 26 Eaton ML. Chronic hypervitaminosis A. Am ] Hosp Pharm 1978; 35: 1099-102. We are grateful to Dr T J Morris for allowing us to include one of 27 Carpenter TO, Pettifor JM, Russell RM, et al. Severe hyper- his patients in our study. vitaminosis A in siblings: Evidence of variable tolerance to intake. IPaediatr 1987; 111: 507-12. 28 Leo MA, Arai Sato M, Lieber CS. Hepatotoxicity of moderate vitamin A supplementation in the rat. Gastroenterology 1982; 1 Di Sant Agnese PA, Talamo RC. Pathogenesis ofcystic fibrosis 82: 194-205. of the pancreas. New EnglJ' Med 1967; 277: 1399-408. 29 Weber FL Jr, Mitchell GE Jr, Powell DE, Peiser BJ, Banwell 2 Goodchild MC, Dodge JA. Cystic fibrosis. Manual ofdiagnosis JG. Reversible hepatoxicity of moderate vitamin A accumu- and management. 2nd ed. Eastbourne, Bailliere Tindall. lation in a protein deficient patient. Gastroenterology 1982; 1985: 12-5. 82: 118-23. 3 Congden PJ, Bruce G, Rothburn MM, Clarke PCN, Little- 30 Pasantes-Morales H, Wright CE, Gaull GE. Protective effect wood JM, Kelleher J, et al. Vitamin status in treated patients oftaurine, zinc and tocopherol on retinol-induced damage in with cystic fibrosis. Arch Dis Child 1981; 56: 708-14. human lymphoblastoid cells.]7 Nutr 1984; 114: 2256-61. 4 Underwood BA, Denning CR. Blood and liver concentrations 31 Mallia AK, Smith JE, Goodman DS. Metabolism of retinol- of vitamins A and E in children with Cystic Fibrosis. binding protein and vitamin A during hypervitaminosis A in PaediatrRes 1972; 6: 26-31. the rat. ] Lipid Res 1975; 16: 180-8.